Fluke Power Quality Troubleshooting

Power Quality Troubleshooting
Introduction
Table of Contents
Page
Safety ............................................................ 2
Getting Started ............................................... 3
First Steps ..................................................... 3
Part I: Facility Distribution System

Section 1: Receptacle Branch Circuit ................... 4
Section 2: Service Panels ...................................... 8
Section 3: Transformers ....................................... 10
Section 4: Electrical Noise and Transients ........ 16
Section 5: Lightning Protection .......................... 21
Part II: Three-Phase Loads

Section 6: Polyphase Induction Motors ............. 22
Section 7: AC Adjustable Speed Drives ............. 25
Section 8: Commercial Lighting .......................... 30
A Lineup of Power Quality Culprits ............... 31
Test Tools ....................................... Back cover
International Safety
Standards for Test Tools
Overvoltage
Category
Summary Description
CAT IV*
Three-phase at utility connection, any

outdoors conductors (under 1000V)
CAT III
Three-phase distribution (under 1000V),

including single-phase commercial lighting
and distribution panels
CAT II
Single-phase receptacle connected loads
CAT I
Electronic
*CAT IV product specifications are not yet defined in the standard.
2 Fluke Corporation Power Quality Troubleshooting
While weve enjoyed enormous

benefits from the evolution of
solid state technology, the fact
is that the microelectronics at
the heart of that technology
requires clean power. Faster
speeds and lower voltages
mean that there is less and less
tolerance for anything less than
quality power.
Power Quality (PQ) covers a
wide range of issues, from voltage disturbances like sags,
swells, outages and transients,
to current harmonics, to performance wiring and grounding.
The symptoms of poor PQ include intermittent lock-ups and
resets, corrupted data, premature equipment failure, overheating of components for no
apparent cause, etc. The ultimate cost is in downtime,
decreased productivity and
frustrated personnel.
This application note gives
you information on how to
troubleshoot PQ problems. It
also gives you information on
how to start fixing those
IEC 61010 establishes international safety requirements for

low voltage (1000V or less)
electrical equipment for measurement, control and laboratory use. The low voltage power
distribution system is divided
into four categories, based on
the proximity to the power
source. Within each category
are voltage listings1000V,
600V, 300V, etc.
The key concept to understand is that you should use a
meter rated to the highest category, as well as the highest
voltage, that you might be
working in. For PQ troubleshooters, that means a meter
rated to CAT III-600V or CAT III1000V (the specifications for
CAT IV have not yet been defined by IEC). We recommend
that you do not use CAT II rated
meters, scopes or test leads and
probes on CAT III circuits. The
CAT ratings should be marked
near the voltage inputs of the
instrument. Meters designed to
IEC 348, the previous standard,
will typically not meet the more
stringent safety specs of IEC
61010 CAT III-600/1000V.
problems. But before grabbing

that meter, please read the
following cautionary notes:
1. Suggested measurements
should only be made by
qualified personnel who have
been trained to make these
measurements in a safe
manner, using proper
procedures and test tools
rated for work on electrical
power circuits.
2. To the best of our knowledge,
recommended solutions are
consistent with the National
Electric Code (NEC), but in
any case, NEC requirements
must not be violated.
3. We have tried to make the
information accurate and
current, but it is not intended
to be a substitute for the
specialized knowledge and
experience of professional
power quality practitioners.
What this application note offers
is a starter kit, not the final
word on PQ troubleshooting.
IEC 61010 requires increased

protection against the hazards
of transient overvoltages. Transients can cause an arc-over
inside an inadequately protected meter. When that arcover occurs in a high energy
environment, such as a threephase feeder circuit, the result
can be a dangerous arc blast.
The potential exists for serious
harm to personnel as well as
damage to the meter. For more
information, see the Fluke application note ABCs of Multimeter
Safety (document number
B0317UEN) and the Fluke video
The ABCs of Digital Multimeter
Safety (P/N 609104).
Independent Testing and

Certification
Manufacturers can self-certify
that they meet IEC 61010 specs,
but there are obvious pitfalls for
the end-user in self-certification.
Certification by an independent
testing lab provides assurance
that the meter meets IEC requirements. Look for a symbol
and listing number of an independent testing lab such as UL,
CSA, TV, VDE, etc. UL 3111, for
example, is based on IEC 61010.
Getting Started
Lighting Panel
Lighting
Load
Utility
XFMR
MV/480Y
Switch Gear
Motor
M.C.C.
PF
Correction
Capacitor
ASD
Induction
Motor
Recept.
XFMR
480/208Y
Recept.
L.C.
Receptacle
Load
Simplified electrical distribution system typical of commercial and industrial facilities.
Start at the scene

of the crime
To troubleshoot PQ problems,
one approach is to start as close
to the victim load as possible.
The victim load is the sensitive load, typically electronic,
that is somehow malfunctioning. Poor PQ is suspected, but
part of your job is to isolate PQ
as a cause from other possible
causes (hardware, software?).
Like any detective, you should
start at the scene of the crime.
This bottom-up approach can
take you a long way. It relies
on making use of a sharp eye
and on taking some basic
measurements.
An alternative is to start at
the service entrance, using a
three-phase monitor, and work
back to the victim load. This is
most useful if the problems
originate with the utility. Yet
survey after survey has concluded that the great majority
of PQ problems originate in the
facility. In fact, as a general rule,
PQ is best at the service entrance (connection to utility)
and deteriorates as you move
downstream through the distribution system. Thats because
the facilitys own loads are
causing the problems. Another
illuminating fact is that 75% of
PQ problems are related to wiring and grounding problems!
For this reason, many PQ
authorities recommend that a
logical troubleshooting flow is
to first diagnose the electrical

infrastructure of the building,
then monitor if necessary. Our
bottom-up troubleshooting procedure is designed to help you
do this detective work.
First steps
1. Make a map: Obtain or
create a current one-line
Its tough to diagnose PQ problems without having a working
knowledge of the site being
investigated. You can start by
locating or reconstructing a
one-line diagram of the site.
The one-line will identify the
ac power sources and the loads
they serve. The as built oneline, the one with red-lines, is
the one you want.
If you work on-site, the map
might already exist in your
head, but it will be a big help to
everyone, including yourself, if
its on paper. If youre coming to
a work site for the first time,
getting an up-to-date one-line
means identifying new loads or
other recent changes in the system. Why go to this effort? Systems are dynamic; they change
over time, often in unplanned
and haphazard ways. Furthermore, while some problems are
local in origin and effect, there
are many problems that result
from interactions between one
part of the system and another.
Your job is to understand these

system interactions. The more
complete your documentation,
the better off youll be.
Its true, however, that the
sites that need the most help
are the ones least likely to have
a good record of whats going
on in their system. Many a consultant has earned his fee by
upgrading the documentation
handed him with what actually
exists on-site. So the simple rule
is, at this point in the investigation, do the best you can to get
good documentation, but dont
count on it being available.
2. Do a walk around
of the site
Sometimes a visual inspection
will offer immediate clues:
A transformer thats much
too hot
Wiring or connections discolored from heat
Receptacles with extension
strips daisy-chained to extension strips
Signal wiring running in the
same trays as power cables
Extra neutral-ground bonds
in sub-panels.
Grounding conductors connected to pipes that end in
mid-air.
At a minimum, you will get a
sense of how the facility is
wired and what the typical
loads are.
3. Interview affected
personnel and keep
an incident log
Interview the people operating
the affected equipment. You will
get a description of the problem
and often turn up unexpected
clues. Its also good practice to
keep a record of when problems
happen and what the symptoms
are. This is most important for
problems that are intermittent.
The goal is to find some pattern
that helps correlate the occurrence of the problem in the
victim load to a simultaneous
event elsewhere. Logically, this
trouble-logging is the responsibility of the operator closest to
the affected equipment.
Power Quality Troubleshooting Fluke Corporation 3
Part I: Facility Distribution System
Section 1
Receptacle Branch Circuit
Many PQ problems show up at
the branch circuit level. Theres
a simple reason for this: thats
where most of the sensitive
loads (and sensitive employees)
are located. Its also the end of
the line of the electrical system, and the place where shortcomings cant be hidden. Lets
assume youve been called in
to solve the problem. Youve
already talked to the people involved, have a rough idea of the
symptoms (equipment lock-ups,
intermittent resets or crashes,
etc.) and as much sense of the
timing and history of the problems as you can get. So its time
to gather hard evidence: its
time to take measurements.
Our primary focus with
troubleshooting at the receptacle level is to determine if the
Line-Neutral (L-N) voltage available is of sufficient stability and
amplitude to supply the needs
of the load(s).
Figure 1.1 Flat-topped voltage at receptacle.
Our other measurements will

tell us whether this flat-topping
is excessive.
2. Peak voltage
The peak value is critical to
electronic loads because the
electronic power supply charges
its internal capacitors to the
peak value of the line voltage.
If the peak is too low, it affects
the ability of the caps to charge
fully and the ability of the
power supply to ride through
momentary dips in the line voltage. For an RMS voltage of
115V, the peak value would be
1.414 x 115V = 162.6V, if the
waveform were a sine wave.
However, as we just saw from
the flat-topped waveform, what
we have is far from a sine wave
and will have a lower peak
value.
which corresponds to the effective heating value. Equipment is

rated in RMS, not peak, because
their main limitation has to do
with heat dissipation.
RMS voltage can be too high
or too low, but it is usually the
low voltage that causes problems. Low RMS voltage combined with flat-topping (low
peak) is a deadly combination
for sensitive loads.
Voltage drop is a function
of both the loading of the circuit
and the source impedance,
which in effect means the
length and diameter (gauge)
of the wire run. The NEC (21019.a, FPN No. 4) recommends a
limit of a 3% voltage drop from
the branch circuit breaker to
the farthest outlet, and a total
voltage drop of less than 5% including the feeder and branch
circuit.
4. Recording (short-term)
The limitation of the above
measurement is that it is static.
Many loads require more current, usually referred to as inrush current, when they are first
turned on. This momentary high
Measurement
current may cause a momentary
low voltage (sag) because of the
1. Waveform
additional IR drop through the
The waveform gives us quick
conductors. Such sags are often
snapshot information. An ideal
caused by loads drawing inrush
waveform would be a sine
currents on the same branch
wave. In this case, (see Fig 1.1)
circuit, or on the same
the voltage waveform is flatpanelboard.
topped, which is typical of a
You can measure a worstbuilding with many non-linear
case sag of 100 ms or more
3.
RMS
voltage
loads such as computers and
(about 6 cycles at 60 Hz) by usNominal line voltage is measured ing the MIN MAX function of the
other office equipment (see
in RMS (root-mean-square)
Flat-topped voltage, page 5).
Fluke 87 while energizing the
load. What if you want to know
if there are recurring sags? The
Table 1.1 Measurements on receptacle branch circuits.
Sags & Swells trending feature
Voltage Measurements
Look for
Instrument
of the Fluke 43 Power Quality
1. Waveform
Snapshot of severity of
43 PQ Analyzer
Analyzer will continuously capvoltage distortion
41B Harmonics Analyzer
ture sags of as little as single
2. Peak voltage
Excessive flat-topping
43 PQA, 41B
cycle duration (17 ms). A four87 DMM (Peak MIN MAX)
minute to a one-hour recording
3. RMS voltage
Low rms (steady-state low
43 PQA (Sags/Swells)
time (i.e., anywhere from a
rms or intermittent/cyclical
41B (MIN MAX)
single cup of coffee to a lunch
sags)
87 DMM (MIN MAX)
break) may be enough to tell
4. Recording (short-term)
Sags, swells, interruptions
43 PQA
you if there are recurring sags
while troubleshooter remains (Sags/Swells or Transients)
and swells.
on-site (4 minutes to 1 hour
typical recording time)
5. Recording (long-term)
Up to 4,000 sags, swells,

outages, transients
VR101S
6. Neutral-ground
N-G voltage too high

(or close to zero)
43 PQA, 87 DMM
Flat-topped voltage
The flat-topped waveform is
typical of the voltage in a commercial building with computer
loads. What causes flat-topping?
The utility supplies ac power,
but electronic equipment runs
on dc power. The conversion
of ac into dc is done by a power
supply. The PS has a diode
bridge which turns ac into pulsating dc, which then charges a
capacitor. As the load draws the
cap down, the cap recharges.
However, the cap only takes
power from the peak of the
wave to replenish itself, since
thats the only time the supplied
voltage is higher than its own
voltage. The cap ends up drawing current in pulses at each
half-cycle peak of the supplied
voltage. This is happening with
virtually all the electronic loads
on the circuit. Now that we see
what the loads are demanding
from the source, lets take a look
at what the source can supply.
If the source were perfectly
stiff, meaning that it had an
infinite capacity to supply all
the current that was required,
then there would be no such
thing as flat-topping (or sags
or any voltage distortion). Think
of it this way: if you had all the
money in the world, you
wouldnt get distorted either
when the bills came in. But in
the real world there are practical limits to what a source can
supply. This limit is usually

described by a concept called
source impedance, which is the
total impedance from the point
youre measuring (or the point
where the load is located) back
to the source. There are two
major contributors to this source
impedance. One is the wiring;
the longer the conductor and
the smaller the diameter (higher
gauge), the higher the impedance. The other factor is the
internal impedance of the transformer (or other source equipment). This internal impedance
is simply a way of saying that a
transformer of a given size/rating can only supply so much
current.
The source impedance is
naturally greatest at the end
of a branch circuit, the farthest
point from the source. Thats the
same place where all those
electronic loads are demanding
current at the peak of the wave.
The result is that the voltage
peak tends to get dragged
downin other words, flattopped. Maybe youve felt the
same way when all the bills
come in at the same time of the
month. The more loads there
are (the more the bills), the
greater the flat-topping. Also,
the greater the source impedance (the less the cash), the
greater the flat-topping.
Diode-Capacitor
Input Circuit
Switching
Power Supply
5. Recording (long-term)
For longer term recording, the
VR101S Voltage Event Recorders will record sags, swells, outages, transients and frequency
deviations while plugged into
the outlet (see Recording at the
Receptacle Outlet, page 7). The
device can be left on-site, unattended, for days and weeks, all
the time catching intermittent
events (4000 event buffer). Now
you can see why its so important to ask the user to keep a
troubleshooting log: correlation
of equipment malfunction with
voltage events is hard evidence
of a PQ problem.
6. Neutral-to-ground voltage
Lets say that you make a
simple L-N measurement at the
outlet and get a low reading.
You cant tell if the reading is
low because the feeder voltage
is low (at the subpanel), or if the
branch circuit is overloaded. You
could try to measure the voltage
at the panel, but its not always
easy to tell which panel feeds
the outlet youre measuring and
its also sometimes inconvenient
to access a panel.
N-G voltage is often an easier
way of measuring the loading
on a circuit. As the current travels through the circuit, there is a
certain amount of voltage drop
in the hot conductor and in the
neutral conductor. The drop on
the hot and neutral conductors
will be the same if they are the
same gauge and length. The total voltage drop on both conductors is subtracted from the
source voltage and is that much
less voltage available to the
load. The greater the load, the
greater the current, the greater
the N-G voltage.
Think of N-G voltage as the
mirror of L-N voltage: if L-N
voltage is low, that will show
up as a higher N-G voltage
(see Fig. 1.4).
Figure 1.2 Flat-topped voltage.
1
2
3
Receptacle N-G Voltage

Measurement Notes
1. A rule-of-thumb used by many in the industry
is that N-G voltage of 2V or less at the receptacle is okay, while a few volts or more
indicates overloading; 5V is seen as the upper
limit. Theres obviously some room for judgment
in this measurement.
2. A high reading could indicate a shared branch
neutral, i.e., a neutral shared between more
than one branch circuit. This shared neutral
simply increases the opportunities for overloading as well as for one circuit to affect another.
3. A certain amount of N-G voltage is normal in a
loaded circuit. If the reading is stable at close
to 0V, suspect an illegal N-G bond in the
receptacle (often due to loose strands of the
neutral touching some ground point) or at the
subpanel. Any N-G bonds other than those at
the transformer source (and/or main panel)
should be removed to prevent return currents
flowing through the ground conductors.
4. If N-G voltage is low at the receptacle, youre in
good shape (see Measurement Note #3 for the
exception to the rule). If its high, then you still
have to determine if the problem is mainly at
the branch circuit level, or mainly at the panel
level. Remember, assuming theres no illegal
N-G bond in intervening panels or receptacles,
your ground test lead goes all the way back to
the source, so youre reading voltage drops all
the way to the source.
N-G voltage exists because

of the IR drop of the current
travelling through the neutral
back to the N-G bond. If the
system is correctly wired, there
should be no N-G bond except
at the source transformer (at
what the NEC calls the source of
the Separately Derived System,
or SDS, which is usually a
transformer). Under this situation, the ground conductor
should have virtually no current
and therefore no IR drop on it.
In effect, the ground wire is
available as a long test lead
back to the N-G bond.
To panel
ground buss
3
To transformer
X0
N-G voltage
Figure 1.3. Neutral-to-ground voltage increases with shared neutrals.
200 ft
60m
50 ft
15m
Laser
Printer
Copier
Figure 1.4 Neutral-ground voltage increases as load current goes up.
Shared neutrals
Some buildings are wired so
that two or three phases share a
single neutral. The original idea
was to duplicate on the branch
circuit level the four wire (three
phases and a neutral) wiring of
panelboards. Theoretically, only
the unbalanced current will return on the neutral. This allows
one neutral to do the work for
three phases. This wiring shortcut quickly became a dead-end
with the growth of single-phase
non-linear loads. The problem
is that zero sequence current
from nonlinear loads, primarily

third harmonic, will add up
arithmetically and return on the
neutral. In addition to being
a potential safety problem
because of overheating of an
undersized neutral, the extra
neutral current creates a higher
N-G voltage. Remember that
this N-G voltage subtracts from
the L-N voltage available to the
load. If youre starting to feel
that shared neutrals are one of
the worst ideas that ever got
translated to copper, youre not
alone.
Solutions
Performance Wiring vs.
Code Minimum
Any experienced PQ troubleshooter will tell you that the
first place to look for most problems is in the building wiring
system (including its grounding
system). Quality power depends
on quality wiring; the term the
industry uses is performance
wiring (See Table 1.2). The basic intent of performance wiring
is to maintain or restore L-N
voltage to the load. There is a
distinction between performance wiring and code minimum wiring. The NEC sets the
absolute minimum requirements
for a wiring job and is primarily
concerned with fire prevention
and personnel safety. The NEC
should, of course, never be violated, but it is also important to

understand that the Codes objective is not to establish standards to achieve power quality.
However, many facilities are
finding that it pays to take the
extra step and install or even
retrofit a performance wiring
job. As one veteran said, If
every building were performance wired, Id be out of
business. . . But theres no fear
of that happening.
Power conditioning
There are also situations where
receptacle-installed power conditioning devices are a good solution, either as a complement
to the wiring changes or as an
economically viable alternative
to some wiring changes.
200 ft
60m
Recording at the
receptacle outlet
By monitoring voltage events at
the receptacle, you can see exactly the same voltage that the
sensitive load sees.
The VR101 is plugged into
an outlet, and can record up
to 4000 events, including:
Voltage sags and swells (rms)
Outages
Transients (L-N and N-G)
with peak values
Frequency deviation
Events are identified by type,
real-time stamp, and duration.
VR101S operation
Set up
Use EventView software to
configure the device. The unit
comes with default thresholds,
but users can enter new thresholds. An optical wand, supplied
with the software, transfers
new configurations to the VR101.
Plug in
The VR101 is left on-site for as
long as needed. No computer
connection is necessary. It
draws power from the line and
in the event of outages, a builtin battery saves data.
50 ft
15m
Laser
Printer
Download
The VR101 is taken to the
computer. The optical wand
retrieves its data.
Copier
Figure 1.5 Performance wiring.
Table 1.2 Suggestions for performance wiring of branch circuits.

Recommendation
Reason
Check for loose connections.
Its easy to overlook the obvious.
Eliminate shared neutrals. In new

installations, pull individual neutrals
for each branch circuit.
Minimize load interaction and source

impedance.
Limit the number of receptacles per

branch circuit to three.
Minimize loading and load interaction.
Limit length of 120V branch circuits

to 50 ft. (15m).
Minimize source impedance.
Analyze
Events are displayed in spreadsheet format in EventView
software. Charts, graphs and
waveform graphics are also
provided for report generation.
Install dedicated branch circuits for all

Keep victim loads and culprit loads separated.
laser printers and copy machines. Dedicated Conduit prevents coupling between circuits.
circuits should be run in their own conduit.
Install a green wire ground (dont just
depend on the conduit connection).
Maintain a continuous, low impedance

ground.
Label all panels, circuit breakers and

receptacles.
Strictly speaking, this wont improve power

quality, but it will sure make life easier for
the troubleshooter and the installer.
Table 2.1 Service panel measurements.
Section 2
Service Panels
Check-out the service panel
as follows:
Visual inspection
Feeder conductor current test
Neutral conductor current test
(feeder and branch)
Phase-to-neutral voltage test
(feeder and branch)
Neutral-to-ground voltage
test (feeder)
Circuit breaker voltage drop
and current on branch phase
conductors
The service panel is where the
effects of single-phase harmonic
loads are easy to measure. A
true-rms meter ensures accurate
readings of non-linear voltages
and currents (see Why Truerms, page 27).
Visual inspection
Look for an illegal NeutralGround bond in subpanels.

This is a violation of the NEC
as well as of PQ wiring. It is
also extremely common. If an
illegal N-G bond is found in
one panel at a site, it is likely
to be in any number of them.
Who knows why theyre
there: perhaps the installer
was thinking that all panels
are wired like residential ser-
Measurement
Look for
Instrument
1. Feeder phase current
Overloading and balance.
43, 41B, 87 w/80i-400,

True-rms ClampMeter
2. Feeder neutral current
High currents from unbalanced fundamental and

3rd harmonics.
43, 41B for spectrum.

87 to find dominant
frequency.
3. Feeder N-G voltage
High voltage indicates

excessive current, near-zero
indicates possible subpanel
N-G bond.
Same
4. Branch L-N voltage
Low voltage.
Same
5. Branch neutral current
Shared neutrals.
Same
6. Voltage drop across

breaker contacts.
Hot breakers.
Worn contacts. Breakers in

need of replacement.
43, 87
vice panels; or that the

quickest way to reduce N-G
voltage was to install a
jumper, or that the more
grounds the better. In any
case, remove all illegal
N-G bondsno exceptions.
Look for signs of overheating, such as discolored connecting lugs. Loose connections and excessive loading show up as heat. High
levels of harmonic current
that were not accounted for
in the original wire sizing
can also cause overheating.
Infrared sensors are the
preferred method for
non- contact temperature
measurement.
Of particular concern is the
size of the feeder neutral
conductor. It has long been
Line
Neutral
N-G
Bond
Ground
Panel
Sub-Panel
Figure 2.1 Sub-panel N-G bonds cause load return currents to flow on ground conductors.
This causes corrosion of pipes in grounding system as well as noisy grounds.
understood that any fundamental current resulting from

the unbalance of single phase
loads among the three phases
will return on the neutral, but
a relatively recent phenomenon is the third harmonic
(triplen) currents generated by
nonlinear single-phase loads
that all return on the neutral.
The 1996 NEC for the first
time stated that On a 4-wire,
3-phase wye circuit where the
major portion of the load consists of nonlinear loads, there
are harmonic currents present
in the neutral conductor, and
the neutral shall be considered
to be a current-carrying conductor. (Article 310, Notes to
Ampacity Tables of 0 to 2000
Volts, Note 10.c). In effect, this
requires that the neutral conductor at least equal the size of
the phase conductor. This requirement is based on solid research: a 1990 survey of 146
sites nation-wide found that
22.6% of them had neutral
current in excess of 100% of
phase current!
Many experts would recommend that the neutral be
double the size of the phase
conductor.
Check for shared branch
neutrals. Count neutral conductors for branch circuits: if
there are fewer than the phase
conductors, there are shared
neutrals.
Check tightness of conduit
connections, especially if the
conduit is being used exclusively as the grounding conductor (not recommended).
Measurements
1. Feeder phase current
Check each phase to make sure
it is not overloaded. Also check
for excessive unbalance.
2. Feeder neutral current
Measure the feeder neutral
conductor for cumulative neutral current. Third harmonic
currents from all three phases
will add arithmetically in the
neutral.
3. Feeder neutral-to-ground
voltage test
As at the receptacle, excessive
N-G voltage indicates overloading. A N-G voltage at or very
near zero indicates the existence of an illegal N-G bond in a
subpanel.
Circuit breaker
voltage drop
Feeder
phase
current(s)
Feeder
neutral
current
4. Phase-to-neutral
voltage test
Phase-to-neutral voltages are
measured and recorded. They
can be compared with receptacle L-N voltages to measure
voltage drop.
5. Branch neutral current
Measure each branch neutral
for overloading. The neutrals
are measured instead of the
phase conductors because they
might share the return current
of several phase conductors,
yet they are not protected
by breakers.
6. Circuit breaker
voltage drop
The voltage drop across a set
of breaker contacts will give
you a quick measure of the
wear of those contacts. Ideally,
the voltage drop should be
zero. In practice, there will be
some voltage drop in the mV
range, with the exact value being dependent on the load current. As a general rule, the
voltage drop should not exceed
20-100 mV, depending on load.
Replace worn breakers.
Branch
L-N
voltage
Branch neutral current

N-G voltage
Figure 2.2 Panel with clamps, probes to show measurements.
Solutions
Table 2.2 Service panel recommendations.
Recommendation
Reason
Limit length of 208V feeder runs to 120V

subpanels to 200 ft. (65m).
Minimize source impedance and chance

of voltage sags.
Dont cascade (daisy chain) subpanels off

of other subpanels if possible, and
especially if the upstream panel is heavily
loaded or has loads with high inrush
currents.
Upstream loads can cause voltage sags that

will affect all downstream loads.
Install a green wire ground conductor

(dont rely on conduit connections).
Maintain a continuous, low impedance

ground.
Reduce the load on the panel if necessary.
Minimize heat, voltage sags.
Redistribute branch circuit loads to improve

balance of the three phases.
Reduce neutral return current (of the

fundamental current).
Upsize the feeder neutral if necessary, to

accommodate the third harmonic. This can
be done by running another neutral
in parallel.
Prevent overloading and heating of feeder

neutral. Will reduce N-G voltage.
Install 3rd harmonic filter.
Reduce neutral current.
Nonlinear load panel.
Manufacturer-designed for nonlinear loads.
Section 3
Transformers
Transformers are subject to
overheating from harmonic currents. Transformers supplying
non-linear loads should be
checked periodically to verify
operation within acceptable
limits. Transformers are also
critical to the integrity of the
grounding system.
Measurements
1. Transformer loading (kVA)
If the transformer has a fourwire wye secondary, which is
the standard configuration for
commercial single-phase loads,
actual kVA can be easily
determined. (See Figure 3.2)
Connect voltage probes on
Phase 1 and Neutral and
clamp current probe on same
phase. Repeat for Phase 2
and 3.
Read kVA of each phase and
sum all three for total transformer kVA.
Compare actual load kVA

to nameplate kVA rating to
determine % loading.
If the load is balanced, a single
measurement is sufficient. (see
Figure 3.1) Transformers loaded
at less than 50% are generally
safe from overheating. However,
as loads increase, measurements should be made periodically. At some point the
transformer may require derating (see page 15).
Red
Black
Figure 3.3 Harmonic spectrum.
For a balanced load: kVATOTAL = kVA1 3

= (kVA1)(1.732)
Figure 3.1 Measuring transformer load (balanced).
1
Red
kVA1
2
Red
Black
kVA2
Black
N
Black
3
Red
kVA3
Unbalanced load:
kVATOTAL = kVA1 + kVA2 + kVA3
Figure 3.2 Measuring transformer load (unbalanced).
2. Harmonic spectrum
The harmonic spectrum of the
secondary (load) current will
give us an idea of the harmonic
orders and amplitudes present:
In a transformer feeding
single-phase loads, the principal harmonic of concern is
the 3rd. The 3rd will add
arithmetically in the neutral
and circulate in the delta primary of a delta-wye transformer. The good news is
that the delta-wye tends to
isolate the rest of the system
from the 3rd (though not the
5th, 7th or other non-triplen
harmonics). The bad news is
that the transformer pays the
price with additional heat.
In a transformer feeding
three-phase loads which include drives or UPS systems
with 6-pulse converters, the
5th and 7th harmonic will
tend to predominate. Excessive 5th is of particular concern because it is negative
sequence. It will tend to
produce counter-torque and
overheating in polyphase
motors.
Table 3.1 Measurements at the distribution transformer.
3. Total Harmonic Distortion

Check for THD of both voltage
1. kVA
Transformer loading. If loading
43, 41B
and current:
exceeds 50%, check for harmonics
For voltage, THD should
and possible need for derating.
not exceed 5%
2. Harmonic spectrum
Harmonic
orders/amplitudes
present:
Same
3rd harmonic (single-phase loads)

For current, THD should not
5th, 7th (primarily three-phase loads)
exceed 5-20% (Table 3.2)
Resonance of higher order harmonics
IEEE
519 sets limits for harmonEffectiveness
of
harmonic
trap
filters
ics at the PCC (Point of Common

3. THD
Harmonic loading within limits:
Same
Coupling) between the utility
Voltage %THD <5%
Current %THD <5-20% (Table 3.2)
and customer (EN50160 is the
European standard). IEEE 519
4. K-factor
Heating effect on transformer from
Same
harmonic loads
is based on THD measurements
taken at the PCC. Technically,
5. Ground currents
Same
Objectionable ground currents are
not quantified but are prohibited by
True-rms Clamp
the PCC is the primary of the
the NEC.
utility supply transformer (alNeutral-ground
bond
in
place
though there are cases where

ESG
(Electrical
Safety
Ground)
the PCC is at the secondary if

connector to ground electrode
(typically building steel) in place
the secondary feeds a number
of customers). In practice, these
measurements are often made
Before-and-after harmonic at the secondary of the
Harmonic amplitudes
spectrum measurement is ex- customers main transformer,
normally decrease as the
tremely valuable to deterfrequency goes up. If one
since that is the point most easmine if harmonic mitigation
frequency is significantly
ily accessible to all parties (and
techniques, like trap filters,
higher in amplitude than
also since that is generally a
which are tuned to specific
lower frequencies, we can
Low Voltage measurement).
frequencies, are sized propsuspect a resonant condition
Some PQ practitioners have
erly and are working as
at that frequency. If such a
broadened the concept of PCC
expected.
condition is detected, be sure
to include points inside the fato take readings at capacitor Different harmonic frequencility, such as on the feeder syscies affect equipment in dif- tem, where harmonic currents
banks to see if the caps are
ferent ways (see below).
experiencing overcurrent/
being generated from one set
overvoltage conditions.
of loads could affect another set
of loads by causing significant
Harmonic Sequences
voltage distortion. The emphasis
Name
F
2nd
3rd
4th
5th
6th
7th
8th
9th
is on improving in-plant PQ,
rather than on simply not afFrequency
60
120
180
240
300
360
420
480
540
fecting utility PQ.
Sequence
+
0
+
0
+
0
Measurement
Look for
Instrument
Rule: If waveforms are symmetrical, even harmonics disappear.

Effects of Harmonic Sequences
Sequence
Rotation
Effects (from skin effect, eddy currents, etc.)
Positive
Forward
Heating of conductors, circuit breakers, etc.
Negative
Reverse
Heating as above + motor problems
Zero
None
Heating, + add in neutral of 3-phase, 4-wire system
Harmonics are classified as follows:

1. Order or number: Multiple of fundamental, hence, 3rd is three times the
fundamental, or 180 Hz.
2. Odd or even order: Odd harmonics are generated during normal operation of
nonlinear loads. Even harmonics only appear when there is dc in the system.
In power circuits, this only tends to occur when a solid state component(s),
such as a diode or SCR, fails in a converter circuit.
3. Sequence:
Positive sequence. Main effect is overheating.
Negative sequence. Create counter-torque in motors, i.e., will tend
to make motors go backwards, thus causing motor overheating.
Mainly 5th harmonic.
3a. Voltage THD

THD has a long history in the
industry. The underlying con
cept is that harmonic currents
generated by loads will cause
voltage distortion (E=IZ) as they
travel through the system
impedance. This voltage
distortion then becomes the
carrier of harmonics systemwide: if, for example, the
distorted voltage serves a linear
load like a motor, it will then
create harmonic currents in that
linear load. By setting maximum
limits for voltage distortion, we
set limits for the system-wide
impact of harmonics.
Zero sequence. Add in neutral of 3-phase, 4-wire system.

Mainly 3rd harmonic.
Table 3.2 IEEE 519 limits for harmonic currents at the point of common coupling.
(All percentages are % of I L, maximum demand load current.)
Odd Harmonics
SCR=Isc/IL
<11
11-17
17-23
23-35
>35
<20
4.0%
2.0%
1.5%
0.6%
0.3%
TDD
5.0%
20-50
7.0%
3.5%
2.5%
1.0%
0.5%
8.0%
50-100
10.0%
4.5%
4.0%
1.5%
0.7%
12.0%
100-1000
12.0%
5.5%
5.0%
2.0%
1.0%
15.0%
>1000
15.0%
7.0%
6.0%
2.5%
1.4%
20.0%
SC R = Short circuit ratio (Isc/IL)

Isc = Available short circuit current at PCC
I L = Maximum demand load current (rms amps)
TDD = Total demand distortion
Note: IEEE allows these limits to be exceeded for up to one hour per day, while IEC
allows them to be exceeded for up to 5% of the time.
The concept of I L , maximum demand load current, is key to using Table 3.2. For existing
facilities, I L is calculated by averaging the maximum demand current for 12 consecutive
months (information available in billing records). For new installations, I L must be estimated. Transformer rating could be used and would be the most conservative estimate
(i.e., it would result in the lowest SCR), since it assumes that the transformer would be
used at full capacity.
The SCR (Short Circuit Ratio) is

a measure of the electrical size
of the customer in relation to
the utility source. The smaller
the customer (higher SCR), the
less the potential impact on the
utility source and the more generous the harmonic limits. The
larger the customer (smaller
SCR), the more stringent the
limits on harmonic currents.
3c. TDD and THD

TDD (Total Demand Distortion) is
the ratio of the current harmonics to the maximum load (I L).
It differs from THD in that THD
is the ratio of harmonics to the
instantaneous load. Why TDD
instead of THD? Suppose you
were running a light load (using
a small fraction of system caFor equipment manufacturpacity), but those loads were
Voltage distortion, however,
nonlinear. THD would be reladepends on source impedance, ers, IEC 1000-3-2, published
in 1995, is the applicable stan- tively high, but the harmonic
i.e., on system capacity. It was
dard. It specifies maximum
currents actually being generquite possible for the first (or
ated would be low, and the efsecond or third) customer to in- current levels out to the 40th
harmonic. Its expected effective fect on the supply system would
ject significant harmonic curdate is projected to be early
in fact be negligible. So who
rents into the system and not
2001. To certify for CE, a recares? TDD acknowledges this,
cause voltage THD to exceed
and allows harmonic load to be
5%. The entire responsibility for quirement for the European
market, manufacturers will
referenced to the maximum
harmonic mitigation could fall
have to meet this standard. This load: if harmonic load is high
on the last customers unlucky
edict will have a major effect on at maximum load, then we have
enough to push VTHD over
to watch out for the effect on
5%, even if their particular har- power supply design.
For the facility, IEEE 519 is
the supply source. So where
monic load was relatively
the standard (EN 50160 in Eudoes that leave current THD
smallliterally the straw that
rope). The limits set in IEEE 519 as a useful measurement. The
broke the camels back.
for harmonic currents depend
closer the current THD
on the size of the customer
3b. Current THD
reading(s) is taken to conditions
of maximum load, the closer it
To restore some fairness to this relative to the system capacity.
(See Table 3.2.)
approximates TDD.
situation, standards for maximum current harmonics were
added, since current harmonics Table 3.3
were under the control of the
Inspection of Transformer Ground
Explanation
local facility and equipment
Check for N-G bond.
A high impedance N-G bond will cause
manufacturer (remember, harvoltage fluctuation.
monic loads act as generaCheck for grounding conductor and
Fault currents will return to the source via
tors of harmonics). This
integrity of connection to building steel
these connections, so they should be as low
emphasis on the mitigation of
(exothermic weld).
impedance as possible.
current harmonics at the load,
Check for tightness of all
If the conduit is not itself grounded, it will
including the not-too-distant
conduit connections.
tend to act as a choke for higher frequencies
and limit fault current (remember that fault
requirement that the load gencurrents are not just at 60 Hz but have
erate virtually no harmonics,
high-f components).
has become the prevailing
Measure for ground currents on the
Ideally there should be none, but there will
regulatory philosophy. It puts
grounding conductor.
always be some ground current due to
the burden of responsibility on
normal operation or leakage of protective
the local site and on the equipcomponents (MOVs, etc.) connected from phase
or neutral to ground. However, anything above
ment manufacturers.
an amp should be cause for suspicion (there is
no hard and fast rule, but experienced PQ
troubleshooters develop a feel for possible
problems).
A final word on measuring

THD: the one place not to apply
the specs is at the individual
harmonic-generating load. This
will always be a worst-case
distortion and a misleading
reading. This is because as harmonics travel upstream, a certain amount of cancellation
takes place (due to phase relationships which, for practical
purposes, are unpredictable).
Measure at a PCC, or at the
source transformer.
transformer supplying office

loads. The higher ratings tend
to be packaged into PDUs
(Power Distribution Units)
which are specially designed
to supply computer and other
PQ-sensitive installations.
5. Ground currents
Two prime suspects for excessive ground current are illegal
N-G bonds (in subpanels, receptacles or even in equipment) and so-called isolated
ground rods:
4. K-factor
Subpanel N-G bonds create
K-factor is a specific measure of
a parallel path for normal rethe heating effect of harmonics
turn current to return via the
in general and on transformers
grounding conductor. If the
in particular. It differs from the
neutral ever becomes open,
THD calculation in that it emthe equipment safety ground
phasizes the frequency as well
becomes the only return
as the amplitude of the harpath; if this return path is
monic order. This is because
high impedance, dangerous
heating effects increase as the
voltages could develop.
square of the frequency.
(Figure 2.1, page 8.)
A K-4 reading would mean
Separate isolated ground
that the stray loss heating
rods almost always create
effects are four times normal. A
two ground references at
standard transformer is, in efdifferent potentials, which in
fect, a K-1 transformer. As with
turn causes a ground loop
THD, it is misleading to make a
current to circulate in an atK-factor reading at the load or
tempt to equalize those poreceptacle because there will be
tentials. A safety and
a certain amount of upstream
equipment hazard is also
cancellation; transformer K-faccreated: in the case of lighttor is what counts. Once the
ning strikes, surge currents
K-factor is determined, choose
travelling to ground at differthe next higher trade size. Kent earth potentials will
factor rated transformers are
create hazardous potential
available in standard trade sizes
differences. (See page 31.)
of K-4, K-13, K-20, K-30, etc.
K-13 is a common rating for a
480V
Transformer grounding
The proper grounding of the
transformer is critical. (Table
3.3.) NEC Article 250 in general
and 250-26 in particular
address the grounding requirements of the SDS.
A ground reference is established by a grounding connection, typically to building
steel (which, in turn, is required to be bonded to all
cold water pipe, as well as
any and all earth grounding
electrodes). Bonding should
be by exothermic weld, not
clamps that can loosen over
time. The grounding electrode conductor itself should
have as low a high-frequency
impedance as possible (not
least because fault current
has high frequency components). Wide, flat conductors
are preferred to round ones
because they have less inductive reactance at higher
frequencies. For the same
reason, the distance between
the grounding electrode conductor connection to the system (i.e., N-G bond at the
transformer) and the grounding electrode (building steel)
should be as short as possible: in the words of the
Code, as near as practicable
to and preferably in the
same area...
The neutral and ground
should be connected at a
point on the transformer neutral bus. Although permitted,
it is not advisable to make
the N-G bond at the main
panel, in order to maintain
the segregation of normal return currents and any ground
currents. This point at the
transformer is the only point
on the system where N-G
should be bonded.
Grounding electrode nearby,

preferably structural metal
208Y/120V
Neutral
Figure 3.4 Transformer grounding.

Solutions
There are a number of solutions
for transformer-related PQ
problems:
Install additional distribution
transformers (Separately
Derived Systems)
Derate transformers
Install K-rated transformers
Used forced air cooling
1. Separately Derived
System (SDS)
The distribution transformer is
the supply for a Separately
Derived System (SDS), a term
which is defined in the NEC
(Article 100). The key idea is
that the secondary of this transformer is the new source of
power for all its downstream
loads: this is a powerful concept
in developing a PQ distribution
system. The SDS accomplishes
several important objectives, all
beneficial for PQ:
It establishes a new voltage
reference. Transformers have
taps which allow the secondary voltage to be stepped up
or down to compensate for
any voltage drop on the
feeders.
It lowers source impedance
by decreasing, sometimes
drastically, the distance
between the load and the
source. The potential for voltage disturbances, notably
sags, is minimized.
It achieves isolation. Since
there is no electrical connection, only magnetic coupling,
between the primary and
secondary, the SDS isolates
its loads from the rest of the
electrical system. To extend
this isolation to high frequency disturbances, specially constructed isolation
transformers provide a
shield between the primary
and secondary to shunt RF
(radio frequency) noise to
ground. Otherwise, the capacitive coupling between
primary and secondary
would tend to pass these
high-frequency signals right
through.
A new ground reference is

established. Part of the definition of the SDS is that it
has no direct electrical connection, including a solidly
connected grounded circuit
conductor, to supply conductors originating in another
system. (NEC 100) The opportunity exists to segregate
the subsystem served by the
SDS from ground loops and
ground noise upstream from
the SDS, and vice versa.
2. K-rated transformers
Figure 3.5 Typical K-factor in commercial

building.
Harmonics cause heating in

transformers, at a greater rate
than the equivalent fundamental currents would. This is because of their higher frequency.
There are three heating effects
in transformers that increase
with frequency:
Hysteresis. When steel is

magnetized, magnetic dipoles
all line up, so that the North
poles all point one way, the
South poles the other. These
poles switch with the polarity
of the applied current. The
higher the frequency, the
more often the switching
occurs, and, in a process
analogous to the effects of
friction, heat losses increase.
Eddy currents. Alternating
magnetic fields create localized whirlpools of current
that create heat loss. This
effect increases as a square
of the frequency. For example,
a 3rd harmonic current will
have nine times the heating
effect as the same current at
the fundamental.
Skin effect. As frequency
increases, electrons migrate
to the outer surface of the
conductor. More electrons are
using less space, so the effective impedance of the conductor has increased; at the
higher frequency, the conductor behaves as if it were a
lower gauge, lower ampacity,
higher impedance wire.
The industry has responded
with two general solutions to
the effects of harmonics on
transformers: install a K-factor
rated transformer or derate a
standard transformer. Lets look
at pros and cons of the K-factor
approach first. K-factor is a
calculation based on the rms
value, %HD (harmonic distortion) of the harmonic currents,

and the square of the harmonic
order (number). It is not necessary to actually perform the calculation because a harmonic
analyzer will do that for you.
The important thing to understand is that the harmonic order
is squared in the equation and
that is precisely where the
high- frequency heating effects,
like eddy current losses, are
taken into account.
K-rated transformers are designed to minimize and accommodate the heating effects of
harmonics. K-rated transformers
do not eliminate harmonics
(unless additional elements like
filters are added). They accommodate harmonics with techniques such as the use of
a number of smaller, parallel
windings instead of a single
large winding: this gives more
skin for the electrons to travel
on. The primary delta winding
is up-sized to tolerate the circulating third harmonic currents
without overheating. The neutral on the secondary is also
up-sized for third harmonics
(typically sized at twice the
phase ampacity).
Application issues with

K-factor transformers
K-rated transformers have been
widely applied, but there are
certain issues with them. Many
consultants do not see the need
for using transformers with a
rating higher than K-13 although K-20 and higher might
be supplied as part of an integrated Power Distribution Unit
(PDU). Also, early applications
sometimes overlooked the fact
that K-rated transformers necessarily have a lower internal
impedance. Whereas a standard
transformer has an impedance
typically in the 5-6% range,
K-rated transformers can go
as low as 2-3% (lower as the
K-rating increases). In retrofit
situations, where a standard
transformer is being replaced by
a K-rated transformer of equivalent kVA, this may require new
short circuit calculations and
re-sizing of the secondary
overcurrent protective devices.
3. Derating standard
transformers
Some facilities managers use
a 50% derating as a rule-ofthumb for their transformers
serving single-phase, predominantly nonlinear loads. This
means that a 150 kVA transformer would only supply 75
kVA of load. The derating curve,
taken from IEEE 1100-1992
(Emerald Book), shows that a
transformer with 60% of its
loads consisting of SMPS
(switched-mode power supplies), which is certainly possible in a commercial office
building, should in fact be
derated by 50%.
The following is an accepted
method for calculating transformer derating for single-phase
loads only. It is based on the
very reasonable assumption
that in single-phase circuits, the
third harmonic will predominate
and cause the distorted current
waveform to look predictably
peaked.
Use a true-rms meter to make
these current measurements:
1. Measure rms and peak current of each secondary
phase. (Peak refers to the instantaneous peak, not to the
inrush or peak load rms
current).
2. Find the arithmetic average
of the three rms readings and
the three peak currents and
use this average in step 3
(if the load is essentially
balanced, this step is not
necessary).
3. Calculate Xformer Harmonic
Derating Factor:
xHDF = (1.414 * I RMS) / I PEAK
4. Or, since the ratio of Peak/

RMS is defined as Crest
Factor, this equation can
be rewritten as:
xHDF = 1.414 / CF
If your test instrument has
the capability, measure the
CF of each phase directly. If
the load is unbalanced, find
the average of the three
phases and use the average
in the above formula.
Since a sine wave current
waveform has a CF=1.414, it
will have an xHDF=1; there will
be no derating. The more the
3rd harmonic, the higher the
peak, the higher the CF. If the
CF were 2.0, then the
xHDF=1.414 / 2 =.71. A CF=3
gives us an xHDF =.47. A wave
with CF=3 is about as badly
distorted a current waveform
as you can expect to see on
a single-phase distribution
transformer.
(Caution: This method does not apply to
transformers feeding three-phase loads,
where harmonics other than the third tend to
predominate and CF is not useful as a simple
predictor of the amount of distortion. A
calculation for three-phase loads is available
in ANSI/IEEE C57.110. However, there is some
controversy about this calculation since it may
underestimate the mechanical resonant
vibrations that harmonics can cause, and that
accelerate transformer wear above and
beyond the effects of heat alone.)
4. Forced air cooling

If heat is the problem, cooling is
the solution. Break out the fan,
turn it on the transformer and
use forced air cooling. Some
experienced hands figure thats
worth 20-30% on the up side.
In any case, it can only help.
Transformer Capacity (%)

After Derating for
Electronic Load
100
80
60
40
20
0
0
20
40
60
80
100
Switched-Mode Power Supply Load (% of Overall Load)
Figure 3-6 Transformer derating curve (IEEE 1100-1992).

Section 4
Electrical Noise and Transients

Electrical noise is the result of
more or less random electrical
signals getting coupled into circuits where they are unwanted,
i.e., where they disrupt information-carrying signals. Noise occurs on both power and signal
circuits, but generally speaking,
it becomes a problem when it
gets on signal circuits. Signal
and data circuits are particularly
vulnerable to noise because
they operate at fast speeds and
with low voltage levels. The
lower the signal voltage, the
less the amplitude of the noise
voltage that can be tolerated.
The signal-to-noise ratio describes how much noise a circuit can tolerate before the valid
information, the signal, becomes
corrupted.
Noise is one of the more mysterious subjects in PQ, especially
since it must be considered
with its equally mysterious
twin, grounding. To lessen the
mystery, there are two key
concepts to understand:
The first is that electrical

effects do not require direct
connection (such as through
copper conductors) to occur.
For an electrician whos been
trained to size, install and
test wiring, this may not be
intuitive. Yet think of lightning, or of the primary and
secondary of an isolation
transformer, or of the antenna to your radio: theres
no direct, hard-wired connection, but somehow complete electrical circuits are
still happening. The same
electrical rules-of-behavior
are in operation for noise
coupling, as will be
explained below.
The second concept is that
we can no longer stay in the
realm of 60 Hz. One of the
benefits of 60 Hz is that its
a low enough frequency that
power circuits can be treated
(almost) like dc circuits; in
other words, basic Ohms
Law will get you most places
you need to go. But when it
comes to noise, we need to
keep in mind that signal circuits occur at high frequencies, that noise is typically
a broad spectrum of frequencies, and that we need to
consider the frequency-dependent behavior of potential
sources of noise.
Noise
20 - 30V
logic signal
Noise
3 - 5V
logic signal
Figure 4.1 Lower voltage, faster signals increase sensitivity to noise.
Coupling mechanisms
There are four basic mechanisms of noise coupling. It pays
to understand them and how
they differ one from the other
because a lot of the troubleshooters job will be to identify
which coupling effect is dominant in a particular situation.
1. Capacitive coupling
This is often referred to as
electrostatic noise and is a
voltage-based effect. Lightning
discharge is just an extreme
example. Any conductors separated by an insulating material
(including air) constitute a capacitorin other words, capacitance is an inseparable part of
any circuit. The potential for
capacitive coupling increases as
frequency increases (capacitive
reactance, which can be
thought of as the resistance to
capacitive coupling, decreases
with frequency, as can be seen
in the formula: X C = 1/ 2 fC).
2. Inductive coupling
This is magnetic-coupled noise
and is a current-based effect.
Every conductor with current
flowing through it has an associated magnetic field. A changing current can induce current
in another circuit, even if that
circuit is a single loop; in other
words, the source circuit acts as
a transformer primary with the
victim circuit being the secondary. The inductive coupling
effect increases with the following factors: (1) larger current
flow, (2) faster rate of change of
current, (3) proximity of the two
conductors (primary and secondary) and (4) the more the
adjacent conductor resembles a
coil (round diameter as opposed
to flat, or coiled as opposed to
straight).
Here are some examples of
how inductive coupling can
cause noise in power circuits:
B
-
Noise Coupling
-G
N-G
Ground
Figure 4.2 Noise coupling. Ground noise measured as -G or N-G noise.
A transient surge, especially

if it occurs on a high-energy
circuit, causes a very fast
change in current which can
couple into an adjacent conductor. Lightning surges are a
worst case, but common
switching transients or arcing can do the same thing.
If feeder cables are positioned such that there is a
net magnetic field, then currents can be induced into
ground cables that share the
raceway.
It is well known that signal
wires and power conductors
should not be laid parallel to
each other in the same raceway, which would maximize
their inductive coupling, but
instead be separated and
crossed at right angles when
necessary. Input and output
cables should also be isolated
from each other in the same
manner.
Magnetic fields are isolated by
effective shielding. The material
used must be capable of conducting magnetic fields (ferrous
material as opposed to copper).
The reason that a dedicated circuit (hot, neutral, ground)
should be run in its own metal

conduit when possible is that is
in effect magnetically shielded
to minimize inductive coupling
effects.
Both inductive and capacitive
coupling are referred to as near
field effects, since they dominate at short distances and distance decreases their coupling
effects. This helps explain one
of the mysteries of noisehow
slight physical repositioning of
wiring can have such major
effects on coupled noise.
3. Conducted noise
While all coupled noise ends up
as conducted noise, this term is
generally used to refer to noise
that is coupled by a direct, galvanic (metallic) connection. Included in this category are
circuits that have shared conductors (such as shared neutrals
or grounds). Conducted noise
could be high frequency, but
may also be 60 Hz.
These are some common
examples of connections that
put objectionable noise currents
directly onto the ground:
Sub-panels with extra N-G
bonds
Receptacles miswired with

N and G switched
Equipment with internal solid
state protective devices that
have shorted from line
or neutral to ground, or that
have not failed but have normal leakage current. This
leakage current is limited by
UL to 3.5 mA for plug-connected equipment, but there
is no limit for permanently
wired equipment with potentially much higher leakage
currents. (Leakage currents
are easy to identify because
they will disappear when the
device is turned off).
Another common example is
the so-called isolated ground
rod. When it is at a different
earth potential than the
source grounding electrode,
a ground loop current occurs.
This is still conducted noise,
even though the direct connection is through the earth.
Datacom connections that
provide a metallic path from
one terminal to another can
also conduct noise. In the
case of single-ended, unbalanced connections (RS-232),
the connection to terminal
ground is made at each end
of the cable. This offers a
path for ground currents if
the equipment at each end
is referenced to a different
power source with a different
ground.
4. RFI (Radio Frequency

Interference)
RFI ranges from 10 kHz to the
10s of MHz (and higher). At
these frequencies, lengths of
wire start acting like transmitting and receiving antennas.
The culprit circuit acts as a
transmitter and the victim circuit
is acting as a receiving antenna.
RFI, like the other coupling
mechanisms, is a fact of life, but
it can be controlled (not without
some thought and effort, however).
RFI noise reduction employs

a number of strategies:
Fiber optic cable, of course,
is immune to electrical noise.
Shielded cabling (such as
coax cables) attempts to
break the coupling between
the noise and signal.
Balanced circuits (such as
twisted pair) dont break the
coupling, but instead take
advantage of the fact that
the RFI will be coupled into
both conductors (signal and
return). This noise (called
Common Mode noise) is then
subtracted, while the signal
is retained. In effect, the balanced circuit creates a high
impedance for the coupled
noise.
Another example of the
high-impedance-to-noise
approach is the use of RF
chokes. Whether used with
data or power cables, RF
chokes can offer effective
high-frequency impedance
(X L increases with frequency).
A low-impedance path can

be used to shunt away the
noise. This is the principle
behind filtering and the use
of decoupling caps (low impedance to high frequency,
but open at power line frequencies). But a sometimesoverlooked, yet critical, aspect is that the ground path
and plane must be capable
of handling high-frequency
currents. High-frequency
grounding techniques are
used to accomplish this. The
SRG (Signal Reference Grid),
first developed for raised
floor computer room installations, is an effective solution.
It is essentially an equipotential ground plane at highfrequency. (For further information on high-frequency
grounding, see the references listed on the back page.)
A Matter of Life and Death

Sometimes PQ troubleshooting
is a matter of life and death.
Dave was the on-site field
engineer at the hospital. One
day he got a call from a very
concerned nurse in the ER. One
of their patients had died. But
as upsetting as that was, it
wasnt the main source of concern. What was really unusual
was that this particular corpse
had a heartbeat.
Dave soon arrived at the
scene. A quick glance told him
that the dead had not come
back to life. The problem lay
elsewhere. The nurses pointed
out what they had seen, a signal on the EKG indicating a
heartbeat. But there was something unusual about this signal
(above and beyond the fact that
it seemed to be coming from a
dead body). He noticed that the
signal was a 60 Hz sine wave
(slightly flat-topped). A further
look at the signal wires told him
that they had been laid parallel

to the power cord. The coupling
between signal and power
wires caused the 60 Hz Heartbeat on the EKG machine.
The moral of the story is to
always isolate the signal and
power conductorsbefore it
becomes a matter of life
and death.
Signal Grounding
To understand the importance of
clean signal grounds, lets discuss the distinction between
Differential Mode (DM) vs. Common Mode (CM) signals. Imagine a basic two-wire circuit:
supply and return. Any current
that circulates or any voltage
read across a load between
the two wires is called DM
(the terms normal mode, transverse mode and signal mode
are also used). The DM signal is
typically the desired signal (just
like 120V at a receptacle).
Imagine a third conductor, typically a grounding conductor.
Any current that flows now
through the two original conductors and returns on this third
conductor is common to both of
the original conductors. The CM
current is the noise that the
genuine signal has to overcome.
CM is all that extra traffic on the
highway. It could have gotten
there through any of the coupling mechanisms, such as
magnetic field coupling at
power line frequency or RFI at
higher frequencies. The point is
to control or minimize these
ground or CM currents, to make
life easier for the DM currents.
Measurement
CM currents can be measured
with current clamps using the
zero-sequence technique. The
clamp circles the signal pair (or,
in a three-phase circuit, all
three-phase conductors and the
neutral, if any). If signal and return current are equal, their
equal and opposite magnetic
fields cancel. Any current read
must be common mode; in other
words, any current read is current that is not returning on the
signal wires, but via a ground
path. This technique applies to
signal as well as power conductors. For fundamental currents, a
ClampMeter or DMM + clamp
would suffice, but for higher frequencies, a high bandwidth instrument like the Fluke 43
Power Quality Analyzer or
ScopeMeter should be used
with a clamp accessory.
Transients
Cursor moves to display peak Min/Max values.
Real-time stamp. Date:hr:min:sec
Figure 4.3 Fluke 43 can capture and save up

to 40 transients.
Transients should be distinguished from surges. Surges are

a special case of high-energy
transient which result from
lightning strikes (see section 5,
PQ Troubleshooting of Lightning Protection Systems). Voltage transients are lower energy
events, typically caused by
equipment switching.
They are harmful in a
number of ways:
They deteriorate solid state
components. Sometimes a
single high energy transient
will puncture a solid state
junction, sometimes repetitive
low energy transients will accomplish the same thing. For
example, transients which
exceed the PIV (peak inverse
voltage) rating of diodes are
a common cause of diode
failure.
Their high-frequency component (fast rise times) cause
them to be capacitively
coupled into adjoining conductors. If those conductors
are carrying digital logic, that
logic will get trashed. Transients also couple across
transformer windings unless
special shielding is provided.
Fortunately this same high
frequency component causes
transients to be relatively
localized, since they are
damped (attenuated) by the
impedance of the conductors
(inductive reactance increases
with frequency).
Utility capacitor switching

transients are an example of
a commonly-occurring highenergy transient (still by no
means in the class of lightning) that can affect loads at
all levels of the distribution
system. They are a well
known cause of nuisance
tripping of ASDs: they have
enough energy to drive a
transient current into the dc
link of the drive and cause
an overvoltage trip.
Transients can be categorized
by waveform. The first category
is impulsive transients, commonly called spikes, because
a high-frequency spike protrudes from the waveform. The
cap switching transient, on the
other hand, is an oscillatory
transient because a ringing
waveform rides on and distorts
the normal waveform.
It is lower frequency, but
higher energy.
Causes
Transients are unavoidable.
They are created by the fast
switching of relatively high currents. For example, an inductive
load like a motor will create a
kickback spike when it is turned
off. In fact, removing a Wiggy (a
solenoid voltage tester) from a
high-energy circuit can create a
spike of thousands of volts! A
capacitor, on the other hand,
creates a momentary short circuit when its turned on. After
this sudden collapse of the applied voltage, the voltage rebounds and an oscillating wave
occurs. Not all transients are the
same, but as a general statement, load switching causes
transients.
In offices, the laser copier/
printer is a well-recognized
bad guy on the office branch
circuit. It requires an internal
heater to kick in whenever it is
used and every 30 seconds or
so when it is not used. This
constant switching has two
effects: the current surge or inrush can cause repetitive voltage sags; the rapid changes in
current also generate transients
that can affect other loads on
the same branch.
Measurement and recording
Transients can be captured by
DSOs (Digital Storage Oscilloscopes). The Fluke 43 PQ
Analyzer, which includes DSO
functions, has the ability to
capture, store and subsequently
display up to 40 transient
waveforms. Events are tagged
with time and date stamps (real
time stamps). The VR101S Voltage Event Recorder will also
capture transients at the receptacle. Peak voltage and real time
stamps are provided.
Voltage susceptibility profile

The new ITIC profile (Information Technology Industry Council) is based on extensive
research and updates the
CBEMA curve. The CBEMA
curve (Computer Business
Equipment Manufacturers Association, now ITIC) was the original voltage susceptibility profile
for manufacturers of computers
and other sensitive equipment.
Similar curves are being developed for 230V/50Hz equipment
and for adjustable speed drives.
Sensitive equipment should be
able to survive events inside the
curve. Events outside of the
curve could require additional
power conditioning equipment

or other remedial action. A major change in ITIC is that the
ride-through times for outages
as well as the tolerance for sags
have both been increased.
The field troubleshooter must
keep in mind that the profiles
are recommendations and that
a particular piece of equipment
may or may not match the profile. Having said that, the profiles are still useful because,
when recorded events are plotted against them, they give a
general idea of the voltage
quality at a particular site.
500
Percent of nominal voltage (RMS or peak equivalent)
Transient voltage surge

suppressors (TVSS)
Fortunately, transient protection
is not expensive. Virtually all
electronic equipment has (or
should have) some level of protection built in. One commonlyused protective component is
the MOV (metal oxide varistor)
which clips the excess voltage.
TVSS are applied to provide
additional transient protection.
TVSS are low voltage (600V)
devices and are tested and certified to UL 1449. UL 1449 rates
TVSS devices by Grade, Class
and Mode. As an example, the
highest rating for a TVSS would
be Grade A (6000V, 3000A),
Class 1 (let-through voltage of
330V max) and Mode 1 (L-N
suppression). The proper rating
should be chosen based on the
loads protection needs:
A lower Grade might result in
a TVSS that lasts one year instead of ten years. The solid
state components in a TVSS
will themselves deteriorate as
they keep on taking hits from
transients.
A lower Class might permit too
much let-through voltage that
could damage the load. Class
1 is recommended for switch
mode power supplies.
A Mode 2 device would pass
transients to ground, where
they could disrupt electronic
circuit operation.
400
300
Applicable to 120, 120/208, and

120/240 nominal voltages
200
140
120
110
100
90
80
70
110V
Voltage-Tolerance
envelope
90V
40
0
0.001c
0.01c
1s
0.1c
0.5c 1c
1ms
3ms
20ms
10c
100c
0.5s
Duration of disturbance in cycles (c) and seconds (s)
Figure 4.4 ITIC Curve.
1000c
10s
Steady
state
Table 5.1 Inspection of lightning protection system.
Section 5
Lightning Protection
Lightning protection plays a
vital part in the overall power
quality of an installation. Lightning occurrence varies by geography, with Florida being the
lightning capital of the U.S.
Lightning does not have to
score a direct hit to be disruptive. It has so much energy that
it couples surges into conductors, both those exposed to air
and those buried in the ground.
Basic lightning protection has
two main requirements:
1. Effective grounding
A low impedance of the grounding electrode system to earth is
important. But, equally important is that all parts of the
grounding system be bonded
together: all ground electrodes
are bonded (and extraneous
ground rods removed), structural
steel is tied to service entrance
ground, all grounding connections are tight and free of corrosion, etc. This minimizes the
phenomenon called transferred
earth potential, where large
surge currents create large voltage differences between two
ground points with different
impedances to earth. This same
grounding practice is important
for performance reasons, as it
tends to minimize ground loop
currents that circulate in an
attempt to equalize ground
potentials.
Check
Look for
Reason
Surge arrestors
Installed at main service

panel, subpanels and
critical equipment.
To minimize high
frequency impedance,
leads should be short,
with no bends.
Grounding electrode
connections are not loose
or corroded.
Grounding conductor
should not be coiled or
have unnecessary bends.
Grounding electrode
conductors at service
entrance or at SDS
All grounding electrodes

should be effectively bonded
together (<0.1).
Prevent difference in earth

potential between electrodes
in event of lightning.
Separately driven
(isolated) electrode
Electrode and equipment

ground should both be tied
to building steel, and
thereby to the service
entrance ground.
Same as aboveentire
grounding system should be
an equipotential ground
plane for lightning.
Datacom cabling that runs

between buildings
Surge arrestors on datacom

cabling or use of fiber optic
cables.
Datacom cabling run

between buildings can be a
path for surge currents, due
to differences between
building earth potentials.
Lightning protection is covered in a number of standards and codes, including:

NEC: Articles 250 and 280
National Fire Protection Association: NFPA 780
Lightning Protection Institute: LPI-175
UL-96 and UL-96A
2. Surge arrestors
A surge arrestor is a protective
device for limiting surge voltages by discharging or bypassing surge current... (NEC 280).
Since the surge current is bypassed to ground, surge arrestors are only as effective as the
grounding system.
Surge arrestors are sized for
the location where they are installed. Three categories are defined (ANSI/IEEE C62.41-1991).
Load
Surge
Arrestor
Surge
Arrestor
Ensure low impedance

ground to minimize
potential to ground with
lightning induced surges.
Minimize impedance to
high frequency
components of lightning.
Grounding electrode bonding
Sub
Panel
Main
Service
Lightning is high energy

and needs multilevel
protection.
Lightning has high f
components. Shorter leads
have less XL and less
impedance at high f.
Surge
Arrestor
A surge arrestor at an outside

installation is closest to the
lightning event and must absorb the most energy. This is
considered a Category C location (corresponding to CAT IV in
IEC 61010). Category B refers to
feeders and distribution panels
(equivalent to CAT III in IEC
61010), and Category A refers to
receptacle connected surge arrestors (equivalent to CAT II).
Surge arrestor or TVSS

A surge arrestor is there to
protect the insulation and, ultimately, prevent failures that
could lead to fires. It is not
necessarily designed to protect
sensitive equipment. Thats the
job of the TVSS (transient voltage surge suppressor).
Figure 5.1. Surge arrestors installed at service, panel, load.
Part II: Three-Phase Loads
Section 6
Voltage unbalance can be caused by severe load unbalance

but it could just as easily be caused by loose connections
and worn contacts.
Polyphase Induction Motors

About two-thirds of the electric
power in the U.S. is consumed
by motors, with industrial
three-phase motors above 5 HP
(7 kW) being by far the bulk of
that load. They are linear loads
and therefore dont contribute
to harmonics. They are, however, the major contributor to
reduced Displacement Power
Factor, which is a measurement
of the effective use of system
capacity.
Measurements
1. Voltage unbalance
Voltage unbalance should not
exceed 1-2% (unless the motor
is lightly loaded). Why such a
small number? Voltage unbalance has a very large effect on
current unbalance, in the neighborhood of 8:1. In other words,
a voltage unbalance of 1% can
cause current unbalance of 8%.
Current unbalance will cause
the motor to draw more current
than it otherwise would. This in
turn causes more heat and heat
is the enemy of motor life, since
it deteriorates the winding insulation.
Example of voltage unbalance calculation:

%VUNBALANCE = Max deviation from average x 100%
Average (of three phases)
1. Make three phase-to-phase measurements:
A-B = 475V A-C = 471V B-C = 470V
2. Find the average: (475+471+470) 3 = 472V
3. Find the maximum deviation from the average:
This occurs on the A-B phase: 475V-472V = 3V
4. Divide maximum deviation by average to find
% unbalance: 3V/472V <1%
2. Voltage %THD and

harmonic spectrum
Voltage THD should not exceed
5% on any phase. If the voltage
distortion on any phase is excessive, it can cause current unbalance. The usual culprit is the
5th harmonic and therefore the
harmonic spectrum should be
examined for the 5th in particular. The 5th is a negative sequence harmonic which creates
counter-torque in the motor. A
motor fed by a voltage with
high 5th harmonic content will
tend to draw more current than
otherwise. This is a major problem when across-the-line or
soft-start motors share the same
bus with ASDs.
Table 6.1 Measurements at the motor.

Measurement
Look for
1. Voltage unbalance
Unbalance <1%
43, 41B, 87
2. Voltage %THD
%THD <5%
43, 41B
3. Current unbalance
Unbalance <10%
43, 41B
87 w/80i-400
Single phasing
(extreme current
unbalance)
No current on one phase.
4. Loading
5. Inrush current
6. Power factor
1
2
Instrument
Nameplate data on FLA1 and SF2 :

Current < (FLA x SF).
Overloading or extreme underloading.
80% of rated load is optimal.
Same
Inrush causing voltage sag.

Inrush causing nuisance trips.
43
Low Displacement PF.

Large difference between DPF and PF
(Total Power Factor) indicating harmonics.
43, 41B
FLA = Full Load Amps

SF = Service Factor
(If the FLA = 100A and the SF=1.15, the motor can be run at 115A continuously.)
3. Current unbalance
To find current unbalance, measure amps in all three phases.
Do the same calculation as for
voltage unbalance. In general,
current unbalance should not
exceed 10%. However, unbalance can usually be tolerated if
the high leg reading doesnt exceed the nameplate FLA (Full
Load Amps) and SF (Service
Factor). The FLA and Service
Factor are available on the motor nameplate. If the voltage unbalance and the voltage THD
are within limits, high current
unbalance can be an indication
of motor problems, such as
damaged winding insulation or
uneven air gaps.
Current measurement will
also find single-phasing. If a
three-phase motor loses a
phase (perhaps caused by a
blown fuse or loose connection),
it may still try to run single
phase off the remaining two
phases. Since the motor acts
like a constant power device, it
will simply draw additional current in an attempt to provide
sufficient torque. A voltage measurement alone will not necessarily find this condition, since
voltage is induced by the two
powered windings into the
non-powered winding.
Flattopping
Effects of inrush current

1. Inrush causes voltage sags if
the source voltage is not stiff
enough:
Relays and contactor coils
might drop out (typically,
the sag would have to get
as bad as about 70% of
Total harmonic distortion < 5%
normal line voltage); or, if
they hold in, their contacts
Figure 6.1. Voltage distortion.
might chatter (especially if
the additional load causes
4. Loading
5. Inrush
a long-term undervoltage).
Measure current draw of the
Motors which are started
Control circuits might reset
motor. If the motor is at or near across-the-line (as opposed to
or lock up (at 90% and
its FLA rating (times the Service those using soft-starts or drives)
below).
Factor multiplier), it will be
draw a current inrush, also
Drives might trip off-line
more sensitive to the additional called locked rotor current. This
(undervoltage trip).
heating from harmonics, as well inrush tapers off to normal run2.
High
peak demand periods,
as current unbalance. A motor
ning current as the motor comes
which may cause higher
that is only lightly loaded is
up to speed.
bills.
usually safe from overheating.
Older motors draw an inrush 3. utility
On the other hand, its efficiency
Cycling loads can cause periof typically 500-600% of the
and DPF are both less than opodic sags, which might show
running current. Newer entimal. Most motors reach maxiup as flickering lights.
ergy efficient designs draw
mum efficiency at 60%-80% of
brief inrushes as high as
4. If the motor is required to
full load rating. Displacement
1200% of running current, a
start up a high torque load,
Power Factor is maximum at
direct result of the lower imthe inrush can be relatively
rated load (including S.F.) and
pedances which help make
prolonged (e.g., 10 to 20
drops off, especially at less than
them more energy efficient
seconds or more) and this
80% of rated load. This leads to
in the first place.
can cause nuisance tripping
the conclusion that, to the deas the overload heaters trip
High
torque,
high
HP
motor
gree a motor load is constant

the motor starter.
loads
require
proportionally
and predictable, 80% of rated
higher
inrush.
load is the most efficient oper6. Power Factor
ating range.
Motor loads started at the
same time will have a cumu- To size PF correction capacitors,
it is necessary to measure the
lative inrush.
DPF (Displacement PF) and
L1
L2
L3
Another source of inrush is UPS Active Power consumption (kW)
and ASD systems with diode
of the motor load. Measurement
converters. They draw inrush
One
of the DPF and kW of a threecurrent as their cap banks first
Open
phase induction motor is exFuse
charge.
plained in the sidebar on the
next page. These measurements
assume that the motor voltage
and current is balanced. Therefore, before undertaking PF correction, first make sure that
voltage and current unbalance
are within limits. Either problem
can shorten motor life and
should take priority over DPF
T2
correction.
T1
T3
M
Single-phasing
Figure 6.3. Inrush current (Model 43).
Figure 6.2. Single-phasing.
Measuring Displacement Power Factor

on 3-phase Induction Motors
Select which of the two methods to use based on
the transformer configuration supplying the motor. Either method will give the same results.
Method #1 is for the grounded-Y source. It is
simple and can be applied in most situations,
since virtually all of the low voltage motors in
commercial and light industrial facilities are fed
from a grounded-Y source. Method #2 is for floating sources sometimes found in heavy industrial
facilities.
Method #1: Grounded-Y source

To check if the source is grounded-Y, measure
voltage of each phase to ground. If the readings
are equal, you can use this measurement method.
Set-up:
1. Clamp the current probe on any phase (with
arrow on clamp pointing towards the motor).
2. Attach the red voltage probe to the same
phase and the black probe to ground (not to
another phase).
Active Power: Read kW and multiply by 3:
kWMOTOR = 3 * kW
Displacement PF: Read DPF.
(Not necessary to measure kVA)
Load
Method #2: Three-wire Source

With floating-Y, floating-delta or grounded-delta,
the voltage will be different for at least one of
the phase-to-ground readings. (For the floating
source, the phase-to-ground voltage is unpredictable,
since it depends on phase to ground capacitance).
Method #2 is known as the Two-Wattmeter method.
Set-up:
1. Connect the black voltage probe to any phase.
2. Connect the red voltage probe and the clamp
(arrow towards the load) together on a second
phase. Record kW1.
3. Move red probe and clamp to the third phase
(do not move the black probe). Record kW2.
Record kVA (kVA of either phase will be more
or less equal if the current unbalance is within
limits).
Active Power: KwMOTOR = kW1 + kW2 (If either
kW reading is negative, as might happen on a
very lightly loaded motor, it would be subtracted
instead of added)
Apparent Power: kVA MOTOR = kVA * 1.73.
DPF = kWMOTOR / kVA MOTOR
Example (Two-Wattmeter):
Measurements:
kW1 = + 1.52
kW2 = + 1.74
kVA = 2.41
Calculations:
KwMOTOR = kW1 + kW2 = (+ 1.52) + (+1.74) = 3.26 kW
kVA MOTOR = kVA * 1.73. = (2.41) (1.73) = 4.17 kVA
DPF = KwMOTOR / kVA MOTOR = 3.26 / 4.17 = 0.78
red
black
red
kW1
kVA
Load
red
black
kW2
black
Power Factor Using the Two-Wattmeter Method:

kWTOTAL = kW1 + kW2
kVATOTAL = kVA1
kWTOTAL
PF =
kVATOTAL
Section 7
PQ Troubleshooting of Adjustable Speed Drives

AC ASDs can be both a source
and a victim of poor PQ (see
Measurement of Adjustable
Speed Drives with Fluke
Meters, document number
G0416UEN, for more information
on drive troubleshooting).
Voltage distortion
If high-voltage distortion shows
up as excessive flat-topping, it
will prevent dc link capacitors
from charging fully and will
diminish the ride-through
capability of the drive. Thus a
voltage sag which would not
ASDs as Victim Loads
normally affect a drive will
cause the drive to trip on
Although ASDs are usually
undervoltage.
depicted as the culprit in the
PQ scenario, there are ways
Improper grounding will affect
in which they can be a victim
the internal control circuits of
load as well.
the drive, with unpredictable
Capacitor switching transients results.
High-energy (relatively lowfrequency) transients that are
ASDs as Culprit Loads
characteristic of utility capacitor A drive can definitely be a culswitching can pass through the prit load and have a major imservice transformer, feeders,
pact on system PQ. But before
and converter front-end of the
we talk of problems, lets put in
drive directly to the dc link bus, a good word for the positive efwhere it will often cause a dc
fects of drives on PQ. First of all,
link overvoltage trip. Input dithey offer built-in soft-start caodes could also be blown out
pabilities. This means there will
by these transients.
Table 7.1 Line-side measurements on ASDs.
Measurement
Look for
Instrument
Voltage waveform
43
Harmonic spectrum
Harmonic orders and amplitudes, before and

after filter application
43, 41B
Displacement PF
For PWM drives, DPF should remain high even

at low speeds (it will typically decrease slightly)
43, 41B
Voltage unbalance
Less than drive manufacturer specs, or current

overload trips can result. The drive may have a
higher limit for unbalance than the motor.
43, 41B
Voltage notching (SCR converters)

Flat-topping
Induction
Motor
M
Line
Reactors
be no inrush current and no

voltage sag effect on the rest of
the system. Secondly, if the
drive is of the PWM type, with a
diode converter front-end, the
Displacement Power Factor is
high (commonly >95% at rated
load) and more or less constant
throughout the range. This
means that drives can reduce
energy usage and correct for
DPF at the same time. Its a
good thing too, because drives
and PF correction capacitors
dont mix (see Power System
Resonance, page 28). Caps are
vulnerable to the higher frequency harmonic currents generated by drives, since their
impedance decreases as frequency increases.
The type of drive has a major
impact on the PQ symptoms, because of the different converter
designs (converters or rectifiers
turn ac to dc and are the first
stage of the drive). There are
two major types of converter
design.
SCR Convertor with Voltage

Source Inverter/Variable
Voltage Inverter (VSI/VVI)
Drives
Commonly called six-step
drives, they use SCRs (Silicon Controlled Rectifiers) in their
converter front-ends (the following discussion applies to
CSI, Current Source Inverter
drives, which also use SCRs).
VSI and CSI drive designs
tended to be applied on larger
drives (>100HP). SCR converters
control the dc link voltage by
switching on (or gating) current flow for a portion of the applied sine wave and switching
off at the zero-crossing points.
Unlike diodes, SCRs require
control circuits for gate firing.
Figure 7.1 Voltage Source Invertor (VSI) ASD
Figure 7.2 Voltage Notching.
For the SCR converter, there are

three main issues that affect
line-side PQ:
Commutation notches. SCR
switching or commutation is
such that there are brief moments when two phases will
both be ON. This causes
what is in effect a momentary short circuit that tends
to collapse the line voltage.
This shows up as notches
on the voltage waveform.
These notches cause both
high V-THD and transients.
The solution is to place a reactor coil or isolation transformer in series with the
drives front end to clean up
both problems.
Displacement Power Factor
declines as drive speed decreases. This is not as serious
a problem as it sounds, because the power requirement
of the drive-motor-load decreases even more.
Harmonic currents, typically
the 5th and 7th, are generated by VSI drives.
Diode Convertor with

Pulse Width Modulation
(PWM) Drives
The other and more common
converter design uses diodes
and is used in the PWM drive.
The diodes require no switching
control circuitry. One of the
main trends in the industry has
been the proliferation of PWM
drives, mainly due to the continued development of fastswitching, efficient IGBTs
(Insulated Gate Bipolar Transistors) used in the inverter section
of the drive (inverters turn dc to
ac). For all practical purposes,
PWM drives are the industry
standard.
For an in-depth discussion of
the efffects of PWM-IGBT drives
on motors, see Measurement of
ASDs with Fluke Meters, document number G0416UEN.
Harmonics solutions
There are a number of solutions

to mitigating drive-generated
harmonics:
Harmonic trap filters (Fig. 7.5)
These are typically LC networks
connected in parallel at the
source of the harmonics (in
other words, at the drive input).
They are tuned to just below
the 5th harmonic (typically 280
Hz) and will tend to sink both
5th and much of the 7th harmonic. Obviously, they must be
sized to the harmonic-generating load.
Phase-shift transformers
This can be as simple as a
delta-wye transformer feeding
one drive(s) and a delta-delta
feeding another drive(s). There
is a 30 degree phase shift effect
Figure 7.3 Typical ASD Harmonic Spectrum
between these two configurations, which effectively results
For the diode converter, the
main PQ issue is harmonics. The in cancellation of harmonics at
the closest upstream PCC (Point
actual harmonic orders being
generated depend on the num- of Common Coupling). The cancellation effect is optimal when
ber of diodes in the front end.
both loads are more or less
For three-phase conversion, a
minimum set of six diodes is re- equal.
Figure 7.4. Pulse Width Modulated (PWM) ASD.
quired. This six-pulse converter will generate 5th and 7th

harmonics. If a 12-pulse converter were used, the 11th and
13th harmonics will be generated instead of the 5th and
6thand, very importantly, for
the same load, the amplitude of
the 11th and 13th would be
considerably less than the 5th
and 6th. Therefore, the THD
would be less. The vast majority
of drives, however, are six-pulse
PWM style, which is one reason
we see so much 5th harmonic
on the system.
12-pulse converter
If the delta-wye/delta-delta are
packaged together (delta primary, delta and wye secondary)
and each secondary feeds one
of two paralleled six-pulse converters, a 12-pulse front-end is
created with all the benefits
mentioned above. 18-pulse
designs are also available. Because of the extra cost, this type
of solution tends to only get
used on high HP loads.
Active filters
This relatively new technology
is based on an elegant concept
using power electronics to solve
the problems created by power
electronics. It senses the instantaneous ac sine wave; it then
actively cancels the harmonics
it detects by generating equal
and opposite polarity harmonics, thus recreating the sine
wave. Commercial packages
might provide voltage regulation as well.
Active PF Correction
Another recent solution is for
manufacturers to offer converter
front ends using fast switching
technology that generates a
minimum amount of harmonics
and has near unity power factor
(both Total PF and DPF).
Line
Load
There is room for discussion

on which approach to harmonic
mitigation might prove most effective and economical in a particular situation. However, what
is often overlooked by the enduser, and what should be clear
from the information in this section, is that the total cost of a
drive system should include
both the cost of the drive itself
and the harmonic mitigation
(whether part of the drive or
installed separately).
Figure 7.5 Harmonic trap filter.
Why True-rms
True-rms test tools are necessary for accurate
measurements of distorted waveforms. For more
information, see the Fluke application note Why
True-rms, document number B0294A.
A comparison of average-responding and true-rms multimeters
Multimeter Reading
Waveform
Description
Average-sensing
DMM
True-rms
DMM
Sine wave
Correct
Correct
Square wave
(flat-top voltage)
10% high
Correct
Current to single
phase diode rectifier
40% low
Correct
Current to 3 phase
diode rectifier
5-30% low
Correct
Power system resonance

Hot vibes can result when harmonics and capacitors get together
Is it possible to install Power
Factor Correction Capacitors
and have PF get worse? It certainly is, and a starting place to
understanding this puzzle lies
in the distinction between Displacement PF (DPF) and Total
Power Factor (PF). The penalty
for not understanding the difference can be blown capacitors
and wasted investment.
Total PF and Displacement PF
are the same in one basic
sense: they are the ratio of Real
Power to Apparent Power, or
Watts to VA. DPF is the classic
concept of power factor. It can
be considered as the power factor at the fundamental frequency. Total Power Factor,
abbreviated to Power Factor
(PF), now includes the effects
of fundamental and of harmonic
currents (it is also referred to as
True PF or Distortion PF,
Fig. 7.7). It follows that with the
presence of harmonics, PF is always lower than DPF and is
also a more accurate description
of total system efficiency than
DPF alone.
Strictly speaking, the term
Power Factor refers to Total PF,
but in practice can also be used
to refer to DPF. Needless to say,
this introduces some confusion
into discussions of power factor.
You have to be clear which one
youre talking about.
Displacement Power
Factor
Lower DPF is caused by motor
loads which introduce the need
for Reactive Power (Volt-Amp
Reactive or VARs). The system
has to have the capacity, measured in Volt-Amps (VA) to supply both VARs and Watts. The
more VARs needed, the larger
the VA requirement and the
smaller the DPF. The cost of
VARs is accounted for in a
power factor penalty charge.
Utilities often levy additional
charges for DPF below a certain
level; the actual number varies
widely, but typical numbers are
0.90 to 0.95.
To reduce VARs caused by
motor loads, power factor correction capacitors are installed.
Upstream system capacity, both
in the plant and at the utility
level, is released and available
for other uses. (Fig. 7.6)
Historically, this has been the
gist of the PF story: a relatively
well-known problem with a
relatively straightforward
solution.
Before: PF = 42%
Harmonics and Capacitors

Harmonics have had a dramatic
impact on our approach to
Power Factor correction. The
motor and capacitor loads described above are all linear and
for all practical purposes generate no harmonics. Non-linear
loads such as ASDS, on the
other hand, do generate harmonic currents.
Take a plant which is stepby-step putting adjustable
speed drives on its motor loads.
ASDs generate significant harmonic currents (5th and 7th on
six-pulse converter drives). Suddenly the fuses on existing PF
correction caps start blowing.
Since these are three-phase
caps, only one of the three fuses
might blow. Now youve got unbalanced currents, possibly unbalanced voltages. The
electrician replaces the fuses.
They blow again. He puts in
larger fuses. Now the fuses survive, but the capacitor blows.
He replaces the capacitor. Same
thing happens. Whats going
on? Harmonics are higher frequency currents. The higher the
frequency, the lower the impedance of a cap (X C = 1/ 2 fC).
The cap acts like a sink for
harmonic currents.
After: PF = 100%
1.4A
3.3A
Active
165 Watts
Reactive
360 VAR
Active
165 Watts
Reactive
360 VAR
Capacitor
60 F
1/6 HP Motor
Figure 7.6 Capacitor corrects Displacement Power Factor (DPF).
1/6 HP Motor
Power System Resonance
which are many times greater

than the exciting current. This
In a worst-case scenario, the
so-called tank circuit can
inductive reactance (XL) of the
severely damage equipment,
transformer and the capacitive
and it will also cause a drop
reactance (XC) of the PF correcin power factor. Perversely, this
tion cap form a parallel resonant resonant condition often apcircuit: XL = XC at a resonant fre- pears only when the system
quency which is the same as or is lightly loaded, because the
close to a harmonic frequency.
damping effect of resistive loads
The harmonic current generated is removed. In other words, we
by the load excites the circuit
have what the audio buffs call
into oscillation. Currents then
a high Q circuit. (Fig. 7.8)
circulate within this circuit
A. System Diagram
B. Equivalent Circuit
XS
XT
XT
XC
XC
XS
Harmonic
source
Figure 7.8 Resonant circuit when XC = (XT + X S)
Imagine coming to work on

a Monday and seeing the insulation on your cables melted off.
How can this happen over a
weekend when there was
hardly any load on the system?
Has Ohms Law been overruled?
Not quite. Your power system
just spent the weekend tanked
out on the Harmonics. It was
quite a party, but now comes
the clean-up.
Start with Harmonics

Mitigation
The correct solution path starts
with measuring and mitigating
the harmonics generated by the
drives. Harmonic trap filters
would generally be called for.
These trap filters are installed
locally on the line side of the
drive. Their effect is very much
like the traditional PF correction
cap, in two senses: they reduce
DPF as well as PF, and also they
localize the circulation of the
problem harmonics (generally
the 5th). Harmonics mitigation
and traditional DPF correction
should be addressed as one
systems issue. In other words,
manage Total PF, not just DPF.
kvar
(nonwork producing)
VA
Harmonics
(nonwork
producing)
kW
(work producing)
Figure 7.7 Total Power Factor increases with harmonics.
Section 8
Troubleshooting Commercial Lighting Loads

Lighting loads are a major load
for many large facilities. Evaluating these circuits is important
for both energy conservation
and power quality. Keep in
mind that commercial lighting
loads are wired single phase,
with the loads connected from
phase to neutral. Typically, the
phase-to-phase voltage is
480V, with the phase-to-neutral voltage at 277V. Measurements must be taken at the
lighting panel, one phase at a
time, since power consumption
and Power Factor could vary on
each phase.
1. Power consumption
Excessive phase unbalance can
cause voltage unbalance, which
in turn can affect three-phase
motor loads.
4. Voltage stability
The sags and swells mode of
the Fluke 43 is especially useful
for recording repetitive voltage
sags which can show up as
flickering lights. Both current
2. Power Factor
and voltage are monitored siBallast with low PF might have multaneously. This helps us to
lower cost-of-purchase but
tell if sags are downstream of
higher cost-of-operation.
the measuring point (loadrelated) or upstream (source3. Total Harmonic Distortion related). For example, if voltage
sags while current swells, a
Current THD should be considered when selecting ballast, es- downstream current inrush
likely caused the sag. If both
pecially if there is a possibility
voltage and current sag, some
of transformer overloading.
event upstream caused
the sags.
Table 8.1 Measurements on commercial lighting loads.
It could be an upstream
load like a motor on a parallel
Measurement
Look for
Instrument
branch circuit which drew
1. Power
Balance among three phases.
43, 41B
down the feeder voltage. Or
consumption (kW)
it could be source voltage2. Power Factor
Magnetic ballast will have low DPF. Electronic
43, 41B
related, for example, a lightning
(DPF and PF)
ballast may have low total PF, although new
generations of ballast often have harmonic
strike or breaker trip/reclosure
mitigation built-in.
on the utility distribution
3. Total Harmonic
Current %THD <20% is desirable.
43, 41B
system.
Distortion (%THD)
4. Voltage Stability
Unstable voltage can cause lights to flicker.
43
480V Phase-to-Phase
2
3
277V Phase-to-Neutral
N
Lighting Panel
Figure 8.1
Figure 8.2 Fluke 43 trends voltage (top) and

current (bottom) simultaneously. Current
swells/inrush caused voltage sags, indicating
that a load downstream from the measurement point is the cause of the disturbance.
A Lineup of Power Quality Culprits

From utility source to receptacle
Lightning
Can be extremely destructive if
proper surge protection is not
installed. It also causes sags
and undervoltages on the utility
line if far away. If close by, it
causes swells and overvoltages.
But in the final analysis, lightning is an act of nature and not
in the same category as the
damage man does to himself.
Utility automatic
breaker reclosure
Causes short duration sags/
outages, but better than the alternative, a longer-term outage.
Utility capacitor switching
Causes a high-energy voltage
disturbance (looks like an oscillating transient riding on the
wave). If the cap bank is near
the facility, this transient can
propagate all through the
building.
Commercial high rises
without enough distribution
transformers
Trying to cut corners in the
wrong places; running 208V
feeder up twenty stories is
not the road to PQ.
Gen-sets not sized for
harmonic loads
Excessive voltage distortion
affects electronic control
circuits. If SCR converter loads
are present, notching can
affect frequency control circuits.
Applying PF correction
capacitors without considering the effects of harmonics
Harmonics and caps dont mix.
Those bulging capacitors are
crying for help.
Inrush currents from high
torque motor loads started
across-the-line
Causes voltage sags if the load
is too large or the source impedance too great. Staggered motor
starts can help.
Undersized neutrals
at panelboard
In the era of the 3rd harmonic,
neutrals can easily carry as
much current or more current
than the phase conductor.
Keeping them undersized leads
to overheated lugs, potential fire
hazards and high N-G voltage.
Running power and signal

cables together
Think of the signal cable as a
single-wire transformer secondary and the power cable as the
primary. The opportunities for
coupling are endless.
Loose conduit connections
and lack of green wire
grounding conductor
Causes open or high impedance
ground circuit. Not good for PQ
or safety.
Shared neutrals on
branch circuits
Causes load interaction and
overloaded neutrals.
Laser printers and copiers
sharing branch circuits with
sensitive loads
Guaranteed periodic voltage
sags and switching transients.
Miswired receptacles
(N-G swapped)
Hard to believe, but they are out
there in quantity. Guaranteed to
put return currents on the
ground conductor and create
a noisy ground.
Data cables connected to
different ground references
at each end
Ouch! Shows up as voltage
between equipment case and
the data cable connector.
Hi-frequency noise
The most effective high frequency grounding technique is
the installation of a Signal Reference Grid (SRG).
And in a class by themselves

Isolated ground rods (below)
Theyre a safety hazard because
the earth is a high impedance
path and will prevent enough
current from flowing to trip the
breaker. They also cause ground
loops; after all, every electron
still has to go back where it
came from. One of the great
mysteries of PQ is how some
manufacturers get away with
insisting that their equipment
warranty is void unless an isolated ground rod is installed. My
auto mechanic was so inspired
by this practice that he now
warranties every car he works
on unless its driven.
Illegal N-G bonds
(Fig. 2.1, page 8)
Guaranteed to put return currents on ground. A common
enough problem that the Union
of PQ Consultants wants to
charge piecework ratessay, a
dollar for every N-G bond
found. Theyd all get rich. Not
only is it a PQ problem, its a
plumbing problem. Circulating
ground currents cause corrosion
of water pipes. That explains
why you can never find the
electrician that put in those N-G
bondsits all being done on
the sly by the plumbers. Guaranteed employment, as if they
needed it.
Load Current
Neutral Return
Current Error
Separately
Derived
System
Line
N-G
Bond
Line
Neutral
Ground
Neutral
Ground
Panel
Earth Ground
Isolated Ground,
Ground Rod, Cold Water Pipe, Etc.
Isolated ground rod can cause ground loops. Common problem with CNC machine tool installations.
A word on test tools

The minimum requirement for
test tools used in PQ troubleshooting are:
True-rms for accurate measurement with harmonics
and distorted waveforms.
In addition, instruments with

recording capability, waveform
display and specialized measurements (such as harmonics,
sags and swells, transient capture, high frequency noise, etc.)
are needed.
CAT III-600V or higher

(CAT III-1000V) safety rating,
which are appropriate for
measurements on power
circuits
The following Fluke test tools are referred to in this application note.
Model
Test Tools
Power
Recording
Real-Time Clock
Fluke 43
Fluke 41B
Fluke VR101S
Fluke 87 Series III
Fluke 36
Power Quality
Analyzer
Harmonic
Analyzer
Voltage Event
Recorder
Digital
Multimeter
ClampMeter
kVA, kW, kVAR, PF, DPF
kVA, kW, kVAR, PF, DPF
TrendPlot, PC logging
PC logging
True-rms volts and

current
True-rms volts and

current
To 51st harmonic
Voltage Transients
20 nanoseconds
with waveform
1 microsecond
event recording
250 microsecond
Peak MIN/MAX
Sags & Swells

(Voltage only)
Single cycle MIN/MAX

with trend
Single cycle
event recording
100 millisecond
MIN/MAX
Sags & Swells

(Simultaneous Voltage
and Current)
Outages

with trend
Event recording with

duration
Documentation,
RS232 Computer
FlukeView
Power Quality Software
FlukeView 41
Software
Motor In-rush Current
Waveform with cursors
Noise
Peak
True-rms
To 31st harmonic
20 MHz scope
Fundamental
References and
acknowledgments
Harmonics
Waveform
4000 voltage events
EC&M Practical Guide to Quality Power for Sensitive Electronic Equipment, 2nd Edition.
Dranetz Field Handbook for
Power Quality Analysis.
Ontario Hydro Power Quality
Reference Guide.
Association of Energy
Engineers: Fundamentals
of Power Quality.
IEEE Std 1100-1992: Recommended Practice for Powering
and Grounding Sensitive Electronic Equipment.
EPRI, Electrotek Concepts, Inc.:
Power Quality for Utility and
Industrial Applications
Power Quality Assurance
Magazine
100 millisecond
MIN/MAX
EventView
Software
MIN/MAX
MAX Hold
Power Quality Troubleshooting

Learn to measure, diagnose and
solve PQ problems.
Part Number: 800919 - NTSC
800927 - PAL
The ABCs of Digital
Multimeter Safety Video Tape
Learn how proper work
procedures and equipment
can protect you from hazards.
Part Number: 609104
For sale at Fluke distributors,
or call the appropriate phone
number listed to the right
Understanding and
Managing Harmonics
Part Number: 609096
Fluke. Keeping your world

up and running.
Fluke Corporation
PO Box 9090, Everett, WA USA 98206
Fluke Europe B.V.
PO Box 1186, 5602 BD
Eindhoven, The Netherlands
For more information call:
U.S.A. (800) 443-5853 or
Fax (425) 356-5116
Europe/M-East (31 40) 2 678 200 or
Fax (31 40) 2 678 222
Canada (905) 890-7600 or
Fax (905) 890-6866
Other countries (425) 356-5500 or
Fax (425) 356-5116
Web access: http://www.fluke.com
1998 Fluke Corporation. All rights reserved.
Printed in U.S.A. 9/98 B0333UEN Rev A
Printed on recycled paper.

Fluke Power Quality Troubleshooting

Transféré par

Informations du document

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Fluke Power Quality Troubleshooting

Transféré par

Droits d'auteur :

Formats disponibles

Power Quality Troubleshooting

Part I: Facility Distribution System

Part II: Three-Phase Loads

Three-phase at utility connection, any

Three-phase distribution (under 1000V),

Single-phase receptacle connected loads

*CAT IV product specifications are not yet defined in the standard.

2 Fluke Corporation Power Quality Troubleshooting

While weve enjoyed enormous

IEC 61010 establishes international safety requirements for

problems. But before grabbing

IEC 61010 requires increased

Independent Testing and

Simplified electrical distribution system typical of commercial and industrial facilities.

Start at the scene

to first diagnose the electrical

Your job is to understand these

Power Quality Troubleshooting Fluke Corporation 3

Part I: Facility Distribution System

Figure 1.1 Flat-topped voltage at receptacle.

Our other measurements will

which corresponds to the effective heating value. Equipment is

Up to 4,000 sags, swells,

N-G voltage too high

4 Fluke Corporation Power Quality Troubleshooting

supply. This limit is usually

Figure 1.2 Flat-topped voltage.

Power Quality Troubleshooting Fluke Corporation 5

Receptacle N-G Voltage

N-G voltage exists because

6 Fluke Corporation Power Quality Troubleshooting

Figure 1.3. Neutral-to-ground voltage increases with shared neutrals.

Figure 1.4 Neutral-ground voltage increases as load current goes up.

from nonlinear loads, primarily

should, of course, never be violated, but it is also important to

Figure 1.5 Performance wiring.

Table 1.2 Suggestions for performance wiring of branch circuits.

Check for loose connections.

Its easy to overlook the obvious.

Eliminate shared neutrals. In new

Minimize load interaction and source

Limit the number of receptacles per

Minimize loading and load interaction.

Limit length of 120V branch circuits

Minimize source impedance.

Install dedicated branch circuits for all

Maintain a continuous, low impedance

Label all panels, circuit breakers and

Strictly speaking, this wont improve power

Power Quality Troubleshooting Fluke Corporation 7

Table 2.1 Service panel measurements.

Look for an illegal NeutralGround bond in subpanels.

1. Feeder phase current

Overloading and balance.

43, 41B, 87 w/80i-400,

2. Feeder neutral current

High currents from unbalanced fundamental and

43, 41B for spectrum.

3. Feeder N-G voltage

High voltage indicates

4. Branch L-N voltage

5. Branch neutral current

6. Voltage drop across

Worn contacts. Breakers in